Individuals with type 1 diabetes mellitus (T1D) who develop an eating disorder are at high risk for diabetes- related medical complications and premature death. Eating disorders (ED) affect a significant subset of T1D patients who often withhold insulin as a means of controlling weight. Yet what accounts for the prevalence of this dangerous comorbidity and how to effectively treat these patients is unknown. Interventions developed for non-diabetic patients have failed to improve key metabolic parameters in T1D patients, suggesting extant models of eating pathology are inadequate for this unique population. Among healthy, non-diabetic populations, eating is largely regulated by the experience of hunger and satiety. Individuals with T1D, on the other hand, must remain constantly cognizant of blood glucose (BG) levels to determine food intake, and errors can have dire consequences. This may generate maladaptive reactions to BG fluctuations (e.g., uncontrolled eating in response to downward trends in BG), and unhealthy attempts to compensate for calories consumed during declining BG and re-establish emotional and behavioral control (e.g., insulin omission). We will study antecedents to ED symptoms among 60 T1D patients with disordered eating using continuous glucose monitors (CGM) time synced to diary entries of psychological state and behavior. Participants will wear the CGM while simultaneously being cued to complete diary entries at random intervals throughout the day and for all eating episodes. Primary aims are as follows: 1) Examine the relationship between fluctuations in BG and ED symptoms among T1D patients, 2) Examine the contribution of diabetes-specific fear and distress to ED pathology, and 3) Describe frequency of insulin omission and psychological factors associated with this behavior among T1D patients with ED. This project is innovative in that it focuses on factors of T1D- ED comorbidity, and specifically on BG decline as a stimulus for disordered eating in diabetes. The methodology of the study capitalizes on the advent of new technologies which allow for the examination of the relationship between fluctuations in BG and eating behavior in ways that were not feasible before. Findings could have a major impact on how we understanding ED in T1D and yield new insights that radically alter treatment. Results may also have implications for disordered eating in type 2 diabetes mellitus, or for ED more generally, one of the leading causes of psychiatric morbidity and mortality among young women. PUBLIC HEALTH RELEVANCE: Individuals with type 1 diabetes (T1D) who develop an eating disorder are at significantly greater risk for diabetes-related medical complications and premature death. Eating disorders (ED) are far more prevalent among individuals with T1D than in the general population. Yet what accounts for the increased prevalence of ED in T1D, and how to effectively treat these patients, is unknown. We will study diabetes-specific factors that contribute to disordered eating and limit treatment effectiveness. The ultimate goal of this work is to improve our understanding of this dangerous comorbidity so that we can develop interventions that improve outcomes for this high risk group.